Dengue continues to extend its range worldwide and has been a significant factor in peacekeeping efforts. Yet, its prevalence and relationship to other important viral infections may make it a good model of flaviviral and hemorrhagic fever infection. A striking feature of infection with the dengue virus is that it is capable of producing a life-threatening illness or a mere laboratory finding associated with no symptoms at all. A potential source for these extremes in disease presentation following infection with dengue viruses is genetic differences in the human host. The 2'-5' oligoadenylate synthetase 1b (OAS 1b) gene of mice makes them resistant to all flaviviral infections. The human orthologs - the OAS 1, 2 & 3 genes - are very similar to the mouse genes. These genes are arranged in tandem, they may be coordinately regulated and probably arose by duplication. They are differentially induced by a variety of interferons and demonstrate many single nucleotide polymorphisms as well as several large variants. This proposal describes a limited case-control study of severe dengue cases from 2002 and 2003 in Salvador, Brazil for associations with these three OAS genes. There were between 40-60 confirmed cases of dengue hemorrhagic fever and thousands of cases of classic dengue in 2002 in this city. A survey of 20 SNP markers across the human OAS genes will be analyzed for allele and haplotype frequencies. The strength of association between alleles or haplotypes and clinical presentation will be assessed as relative risks and corrections will be made for population substructure and for multiple comparisons. In addition, demographic and familial data will be collected to assess the possibility for future familial linkage studies of severe or classic dengue in an R01. Besides excellent epidemiologic and laboratory investigators in Brazil, the project relies on a long history of collaboration between the groups, strong genetic epidemiology support from the Division of Genetic Epidemiology at Case Western Reserve University and serology laboratory support from co-investigators based at the Evandro Chagas Institute, Brazil and Fort Dettrick, MD. The results of this study may help predict populations at risk, may influence vaccine design or may aid in understanding and managing other viral hemorrhagic fevers.